BERLIN – Nvidia is extensively expanding its supply chain even further, with plans to scale its optics capacity fifteenfold in 2027.

The GPU giant’s GTC event in Germany revealed plans with an array of suppliers, including Amkor, Samsung Electro-Mechanics, and Unimicron.

On the networking side, Nvidia confirmed plans with Siliconware Precision Industries (SPIL) to expand manufacturing capacity for optical engines for use in co-packaged optics (CPO) 15x by mid-2027. The vendor even claimed it aims “for another doubling” by the end of 2028.

“Our new Spectrum-6 CPO switch is the building block for scale-out networking for Vera Rubin DSX AI factories. We're very excited about bringing co-packaged optics to the AI data center and working with Spill and others to expand our capacity,” Ian Buck, Nvidia’s VP and GM of hyperscale and high-performance computing, said in a press briefing.

Nvidia first teased its CPO switches in the spring of 2025, with its Spectrum-6 switch line formally unveiled in July this year. While it offers both standard copper and CPO, the company suggests the optical iteration packs 409.6 terabits per second (Tb/s) of bandwidth across 512 ports at 800 Gb/s.

GTC Berlin’s supply chain expansion is not surprising; the Jensen Huang-led company is renowned for its supply chain management magic. In fact, throughout 2026, it’s penned multiple supply chain deals with optics firms like Corning, Lumentum, and Finisar parent Coherent to ensure adequate supply, amid capacity concerns related to electro-absorption modulated lasers (EMLs).

“We are in the midst of the largest infrastructure buildout in human history. Demand is extraordinary. Capacity is the constraint,” said CEO Jensen Huang. “Our supply chain partners are the world’s best manufacturers. I’m grateful they are expanding capacity around the globe to scale production and help us deliver the AI supercomputing systems powering this new industrial revolution.”